Postnatal maturational shift from PKCζ and voltage-gated K⁺ channels to RhoA/Rho kinase in pulmonary vasoconstriction

Abstract

Objective: The neonate is at high risk of developing pulmonary hypertension, which may reflect a misbalance between vasodilator and vasoconstrictor agents. Thromboxane A₂ (TXA₂) is involved in several forms pulmonary hypertension, but the signaling pathways mediating its pulmonary vasoconstrictor responses during postnatal maturation have not been analyzed. We therefore investigated the role of L-type Ca²⁺ channels, protein kinase C (PKC) ζ, voltage-gated K⁺ channels (K_V), and RhoA/Rho kinase in TXA₂-induced pulmonary vasoconstriction during postnatal maturation.

Methods: Changes in contractility and intracellular calcium were analyzed in 1 day (newborn) and 2-week-old piglets' pulmonary arteries (PA). K_V currents were investigated in freshly isolated smooth muscle cells using the whole-cell configuration of the patch clamp technique.

Results: The contractile responses to the TXA₂ mimetic U46619 were similar at both ages but the L-type Ca²⁺ channel blocker nifedipine and a PKCζ pseudosubstrate inhibitor only attenuated the contraction in newborn PA. K_V currents were similarly inhibited by U46619, although their density was dramatically reduced in 2-week-old as compared to newborn PA smooth muscle cells. This was consistent with a greater contraction to the K_V inhibitor, 4-aminopyridine, and with a leftward shift in the increase in intracellular Ca²⁺ by U46619 in newborn versus older animals. On the other hand, the Rho kinase inhibitor Y-27632 induced a stronger inhibitory effect on the contraction induced by U46619 in 2-week-old than in newborn PA and this was accompanied with minor effects on intracellular calcium levels.

Conclusion: TXA₂-induced pulmonary vasoconstriction involves PKCζ-K_V-L-type Ca²⁺ channel and RhoA/Rho kinase signaling pathways, which are downregulated and upregulated, respectively, during postnatal maturation. The different contribution of these pathways could be of relevant importance for the vasodilator therapy choice in the treatment of pulmonary hypertension.

Thromboxane A2, Potassium currents, Protein kinase C, Pulmonary artery

1. Introduction

Thromboxane A₂ (TXA₂), a major product of the constitutive and inducible isoforms of cyclooxygenase (COX-1 and COX-2, respectively) within the pulmonary vascular bed [1], is a potent pulmonary vasoconstrictor. It participates in the control of pulmonary vessel tone under physiological and, especially, pathological situations. In fact, TXA₂ has been involved in several forms of human and experimental pulmonary hypertension including primary [2] and secondary pulmonary hypertension induced by sepsis, endotoxemia, heparin/protamine, leukotriene D₄, microembolism, and ischemia–reperfusion [3–8]. Isoprostanes, a novel class of arachidonic acid metabolites generated by oxygen free radical-mediated peroxidation of arachidonic acid, have also been implicated in pulmonary hypertension [9].

TXA₂ and several isoprostanes bind to specific G protein-coupled receptors (TP receptors) [9,10]. The signaling pathway for TP receptor-induced vasoconstriction involves a variety of protein kinases such as protein kinase C (PKC) and Rho kinase [11–13]. PKC represents a family of several isoforms which can be divided into classic (cPKC: α, βI, βII and γ), novel (nPKC: δ, ε, η and θ), and atypical (aPKC: ζ and λ/ι) isoforms [14,15]. We have recently reported that PKCζ plays a functional role in the vasoconstriction induced by TP receptor activation in rat pulmonary arteries (PA) [11]. The TXA₂ analogue U46619, via PKCζ, inhibited voltage-gated K⁺ channels (K_V channels) leading to membrane depolarization, activation of L-type Ca²⁺ channels, increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) and a contractile response. Because of the key role of K_V channels in the regulation of membrane potential in PA [16], their inhibition is also involved in the vasoconstriction induced by hypoxia and endothelin-1 [17,18]. Furthermore, changes in the expression or function of these channels in PA have been involved in the pathogenesis of primary and anorexigen-induced pulmonary hypertension [17,19–21].

RhoA, a member of the Ras family of small GTP binding proteins, and its target Rho kinase, have been shown to increase Ca²⁺ sensitivity via inhibition of myosin light chain phosphatase (MLCP) and thus play an important role in regulating smooth muscle contractility [22,23]. Both RhoA and Rho kinase are expressed in PA [23–25] and have been involved in hypoxic pulmonary vasoconstriction, hypoxic pulmonary hypertension, and monocrotaline-induced pulmonary hypertension [23,25–28]. In addition, the RhoA/Rho kinase pathway may contribute to TXA₂- and isoprostanes-induced contractions in pulmonary [9,13] and systemic vessels [29]. Finally, this signaling pathway may be important for fetal lung morphogenesis but its continued activation may impair lung development in the neonate [27].

At birth and during the first days of life, as the lung becomes responsible for blood oxygenation, dramatic structural and functional changes occur in the pulmonary circulation to permit the adaptation to extrauterine life [30]. Any alteration in this delicate transition may lead to persistent pulmonary hypertension of the newborn (PPHN) [31]. Thus, a better understanding of the mechanisms involved in triggering pulmonary vasoconstriction during postnatal maturation is a key issue to design more effective pharmacological approaches for the treatment of PPHN. Therefore, in the present study, we have analyzed the signaling pathways mediating TXA₂-induced pulmonary vasoconstriction during postnatal maturation. We have found that even when the vasoconstrictor responses to the TXA₂ mimetic U46619 were similar in newborn and 2-week-old piglets the contribution of the PKCζ-K_V-Ca²⁺ signaling pathway decreases while that of the RhoA–Rho kinase pathway increases with postnatal age.

2. Methods

All experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).

2.1. Reagents

Drugs were obtained from Sigma except nifedipine (Bayer), Y-27632 (Tocris Cookson), fura-2 AM, calphostin C, Gö-6976, PKCζ pseudosubstrate inhibitor, and secondary horseradish peroxidase conjugated antibodies (Calbiochem). Polyclonal rabbit anti-PKCζ and anti-pT410-PKCζ antibodies were from Santa Cruz Biotechnologies and monoclonal anti-PKCλ/ι antibody from Transduction Laboratories, Lexington, KY. NO solutions were prepared as previously described [32].

2.2. Tissue preparation and cell isolation

Third-order branches of the PA (internal diameter, 0.7–1.2 mm) isolated from 17 newborn (3–18 h old, NB) and 20 two-week-old male piglets were dissected in Krebs solution. PA were placed into a nominally calcium-free physiological salt solution (Ca²⁺-free PSS) of composition (in mmol/L): NaCl, 130; KCl, 5; MgCl₂, 1.2; glucose, 10; HEPES, 10 (pH 7.3 with NaOH), cut into small segments (2 × 2 mm) and cells were isolated after enzymatic digestion as previously reported [11].

2.3. Contractile tension recording

Contractile responses in endothelium-denuded PA rings were recorded as previously reported [33,34]. Arteries were firstly stimulated with U46619 (0.1 μmol/L) and once a stable contraction was reached, they were washed with Krebs solution for 60 min. Following this procedure, a concentration-dependent stimulation with U46619 was elicited in the absence (controls) or after 30 min treatment with different drugs. In another set of experiments, the contractile responses to the K_V channel inhibitor 4-aminopyridine (4-AP, 1 mmol/L) were analyzed. The values of the second contraction to either U46619 or 4-AP were expressed as a percentage of the initial response to the TXA₂ analogue.

2.4. Simultaneous [Ca²⁺]_i and tension recording

PA rings were incubated for 2–3 h in Krebs solution containing fura-2 AM (5 μmol/L) and Cremophor EL (0.05%) and then mounted in a fluorimeter, excited through the luminal surface at 340 and 380 nm and the emitted fluorescence was filtered at 505 nm [33,34]. Arteries were initially stimulated with 40 mM KCl, washed in Krebs solution in the absence or presence of different drugs for 30 min, and finally a cumulative concentration–response curve to U46619 was constructed. The U46619-induced changes in the ratio of the emitted fluorescence (F340/F380) and contractile tension were expressed as a percent of the initial response to KCl. In another set of experiments, the effects of several vasodilators were analyzed after stimulation with U46619 (0.1 μmol/L).

2.5. Electrophysiological studies

Membrane currents were measured using the whole-cell configuration of the patch-clamp technique with an Axopatch-200B patch-clamp amplifier (Axon Instruments, Burlingame, CA, USA) as previously described [11]. Cell capacitance was determined by integration of the capacity transient in response to 10 mV hyperpolarization. Currents were normalized for cell capacitance and expressed in pA pF⁻¹. K_V currents (I_K(V)) were recorded under essentially Ca²⁺-free conditions using an external Ca²⁺-free PSS (see above) and a Ca²⁺-free pipette (internal) solution containing (mmol/L): KCl, 110; MgCl₂, 1.2; Na₂ATP, 5; HEPES, 10; EGTA, 10; pH adjusted to 7.3 with KOH. All experiments were performed at room temperature (22–24 °C).

2.6. Western blot analysis, phosphorylation of T410, and cell fractionation

PA were frozen, homogenized and fractionated as described [11]. Western blotting was performed with 20 μg of protein. SDS-PAGE (7.5% acrylamide) electrophoresis was performed using the method of Laemmli in a mini-gel system (Bio-Rad). Samples from newborns and 2-week-old animals were run in parallel. The proteins were transferred to PVDF membranes overnight at 4 °C and incubated with rabbit anti-PKCζ, anti-P-T410-PKCζ or anti-PKCλ/ι primary antibodies and secondary anti-rabbit horseradish peroxidase conjugated antibodies. The bands were visualized by chemiluminiscence (ECL, Amersham). The results were expressed as a percentage of the data of cytosolic fraction from newborn animals.

2.7. Statistical analysis

Data are expressed as means ± S.E.M.; n indicates the number of arteries or cells tested from at least three different animals. The concentration–response curves to U46619 were fitted to a logistic equation and the half-maximum effective concentration expressed as negative log molar (pD₂) and the maximal effect (E_max) were calculated in each experiment. Statistical analysis was performed using Student's t-test for paired observations or one-way ANOVA followed by a Newman–Keuls' test. Differences were considered statistically significant when P<0.05.

3. Results

3.1. Role of L-type Ca²⁺ channels and PKC during maturation

PA were initially contracted by 40 mM KCl (417 ± 69 and 967 ± 112 mg in newborn and 2-week-old piglets, respectively). After washing, concentration–response curves to the TXA₂ analogue U46619 were elicited yielding similar pD₂ (7.26 ± 0.08 and 7.39 ± 0.06) and E_max values (172 ± 22% and 154 ± 16% of the response to KCl, n=5) in newborns and 2-week-old piglets, respectively. In another set of experiments, after an initial stimulation with a submaximal concentration of U46619 (0.1 μmol/L), PA were incubated in the absence or presence of several drugs before constructing the curve to U46619 (Fig. 1). Incubation with the L-type Ca²⁺ channel blocker nifedipine (1 μmol/L) markedly inhibited the vasoconstriction in PA from newborn piglets but had no effects in the older animals. The contraction induced by U46619 in 2-week-old piglets was also resistant to the sarcoplasmic reticulum Ca²⁺ ATPase inhibitor thapsigargin (1 μmol/L, not shown). The role of PKC was analyzed using calphostin C (1 μmol/L), a broad inhibitor of PKC isoforms, Gö-6976 (0.01 μmol/L), which inhibits cPKC isoforms or with a PKCζ pseudosubstrate inhibitor (PKCζ-PI, 10 μmol/L). Similar to nifedipine, calphostin C and PKCζ-PI inhibited the contraction induced by U46619 only in newborn arteries, whereas Gö-6976 had no effect in any age group.

Fig. 1

Effects of (A) the Ca²⁺ channel blocker nifedipine (1 μmol/L), (B) the broad inhibitor of PKCs calphostin C (1 μmol/L) and the inhibitor of classic PKCs Gö-6976 (0.01 μmol/L) and (C) the specific inhibitor of PKCζ (PKCζ-PI, 10 μmol/L) on the vasoconstrictor effects of the TXA₂ analogue U46619 in isolated piglet PA from newborn (NB) and 2-week-old piglets. Data show mean ± S.E.M. (n=5–11). *,** P<0.05 and P<0.01 vs. control, respectively.

3.2. Expression, subcellular distribution, and phosphorylation of PKCζ

In order to test whether the difference in the effect induced by PKCζ-PI with age was due to different PKCζ protein content, we analyzed its expression in PA. An antibody against the C-terminal peptide of PKCζ recognised two bands of approximately 81 and 75 kDa. This antibody cross reacts with the aPKCλ/ι. However, the expression of this aPKC was negligible in PA but present in HEK293 cells (positive control) using a specific anti PKCλ/ι antibody (Fig. 2B). The activity of PKCζ is mostly controlled by its phosphorylation and subcellular compartmentalization (cytosolic vs. particulate fractions). The expression and compartmentalization of PKCζ was not significantly different in both age groups (Fig. 2A and C). Furthermore, an antibody directed towards the phosphorylated activation loop (T410) of PKCζ revealed that phosphorylated PKCζ (pPKCζ) was not significantly different in the two groups (Fig. 2A and D).

Fig. 2

Expression and phosphorylation of PKCζ in newborn (NB) and 2-week-old PA. (A) Representative Western blots of cytosolic (Cyt) and particulate (Part) enriched fractions of homogenates of PA using antibodies directed against the C-terminal domain of PKCζ, the phosphorylated T410 of PKCζ (pPKCζ) and smooth muscle α-actin. (B) Western blots of PKCλ/ι in PA showing negligible protein expression, HEK293 cells were used as positive controls. Quantitative densitometric analysis of PKCζ (C) and pPKCζ (D) in the cytosolic (solid bars) and the particulate fractions (open bars). Data (mean ± S.E.M. of four piglets) are expressed as a percentage of the values in the cytosolic fraction of newborns.

3.3. Effects of U46619 on [Ca²⁺]_i in newborn and 2-week-old PA

In these experiments, we simultaneously measured the changes in [Ca²⁺]_i and contractile force induced by U46619 in fura-2-loaded PA. Under control conditions, the concentration–response curve for the increase in [Ca²⁺]_i induced by U46619 was markedly shifted to the right in 2-week-old as compared to newborn PA (Fig. 3A and B) with pD₂ values of 6.75 ± 0.16 and 7.56 ± 0.16 (P<0.05), respectively. However, the contraction induced by the TXA₂ analogue was similar at both ages (Fig. 3B) with pD₂ values of 7.28 ± 0.06 and 7.26 ± 0.09 (for newborns and 2-week-old PA, respectively (P>0.05). It is interesting to note that contractions to the lower concentrations of U46619 (≥ 0.03 μmol/L) in the 2-week-old animals (but not in newborns) occurred without significant changes in [Ca²⁺]_i. In another set of experiments, incubation with nifedipine (1 μmol/L) or PKCζ-PI (10 μmol/L) nearly abolished the increase in [Ca²⁺]_i induced by the TXA₂ analogue at both ages (Fig. 3C and D) but as shown in Fig. 1, these agents only inhibited the contraction in newborn arteries. Therefore, the nearly complete suppression of [Ca²⁺]_i increase (either with nifedipine or the PKCζ-PI) partially inhibited the contraction induced by U46619 in newborn PA but had no effect in older animals.

Fig. 3

Increases in [Ca²⁺]_i induced by U46619 in fura-2 loaded PA from newborn and 2-week-old piglets. Original recordings of changes in [Ca²⁺]_i elicited by 0.001–3 μmol/L U46619 (A). Average changes (mean ± S.E.M.) in [Ca²⁺]_i and in contractile force induced by U46619 (B). Effects of nifedipine and PKCζ inhibitor peptide in PA from newborns (C) or 2-week-old animals (D). All results are expressed as a percent of an initial response to 40 mM KCl. Data show mean ± S.E.M. (n=4–6).

3.4. Role of K_V channels

A family of K_V currents (I_K(V)) were obtained in PA smooth muscle cells (PASMC) when eliciting depolarizing steps from −60 to +60 mV (Fig. 4A and B). Interestingly, the current amplitudes were much greater in newborns than in 2-week-old PASMC (447 ± 63 and 196 ± 35 pA at +60 mV, respectively, P<0.05, Fig. 4A). Furthermore, the average capacitance was almost twofold lower in newborns than in older animals, so that K_V current density was dramatically reduced in the older animals as compared to newborns (Fig. 4B). In spite of these differences, K_V currents showed similar activation kinetics and voltage–activation curves at the two ages (Fig. 4C and D). Normalized conductance–voltage curves were fitted to a Boltzmann equation yielding midpoints of −1.2 ± 1.1 mV and −1.7 ± 1.5 mV and a slope factors (k) of 11.1 ± 0.4 mV and 12.7 ± 0.9 mV for newborn (n=7) and 2-week-old (n=6) PASMC, respectively (Fig. 4D). U46619 (0.3 μmol/L) caused a significant inhibition of I_K(V) in the whole range of channel activation in both newborn and 2-week-old PASMC (Fig. 4A and E) but had no effect on the voltage–activation curves (Fig. 4D). U46619-induced blockade was not observed in cells dialyzed with the pipette solution containing the PKCζ-PI (0.1 μmol/L, Fig. 4F). The currents were not affected by the large conductance Ca²⁺-activated K⁺ channel inhibitor iberiotoxin (0.1 μmol/L, Fig. 5A). However, the K_V channel inhibitor 4-AP (1 mmol/L) similarly blocked I_K(V) at both ages (Fig. 5B), even when this drug induced a much greater contractile response in newborn (37 ± 9% of the response to U46619 0.1 μmol/L, n=6) than in 2-week-old (7 ± 4%, n=8, P<0.01) piglet PA rings (Fig. 5C). These data indicated that K_V channel blockade induced a more effective pulmonary vasoconstriction in newborn than in older animals and may therefore make a much greater contribution to U46619-induced pulmonary vasoconstriction in newborns than in older animals.

Fig. 5

Inhibitory effects of iberiotoxin (A, 0.1 μmol/L) and 4-AP (B, 1 mmol/L) on K⁺ currents recorded in PASMC from newborn (NB) and 2-week-old piglets. (C) Contractile effects of 4-AP in PA expressed as a percent of a previous response to U46619 (0.1 μmol/L). Data show mean ± S.E.M. (n=4–11). ** P<0.01 vs. NB.

Fig. 4

Effects of U46619 (0.3 μmol/L) on I_K(V) recorded in newborn (NB) and 2-week-old PASMC. (A) Current traces are shown for 200 ms depolarization pulses from −60 mV to +50 mV from a holding potential of −60 mV. (B) Current density and cell capacitance (inset) in PASMC. (C) Activation kinetics of I_K(V). Data show time constants (τ) as a function of test potentials. (D) Activation curve of I_K(V) in the absence or presence of U46619. Conductance (G) was normalized to the maximal conductance (G_max) plotted against the test potential and fitted with the Boltzman equation: G/G_max=1/{1+exp[(V_1/2−V)/k]}, where G is conductance, G_max is maximal conductance, V_1/2 is the voltage for half-maximum activation, V the potential of the test, and k is the slope factor. (E) Voltage-independent I_K(V) blockade induced by U46619. (F) Effect of PKCζ-PI (0.1 μmol/L) on U46619-induced blockade of I_K(V). All data show mean ± S.E.M. (n=5–7). **P<0.01 vs. NB.

3.5. Role of Rho kinase

In order to assess the contribution of Rho kinase to the pulmonary vasoconstriction induced by U46619 during postnatal maturation, PA rings were incubated with the Rho kinase inhibitor Y-27632 (1 μmol/L). The drug inhibited the vasoconstriction induced by the TXA₂ analogue at both ages (Fig. 6A and B), but this effect was much greater in older animals (19.8 ± 3.2% and 40.1 ± 5.4% inhibition of the maximal response to U46619 in newborns and 2-week-old animals, respectively, n=6–8, P<0.05). In another set of experiments, rings were incubated with a combination of Y-27632 (1 μmol/L) plus nifedipine (1 μmol/L) or plus calphostin C (1 μmol/L). The addition of either of these two drugs induced a stronger inhibitory effect on the contraction than that induced by Y-27632 alone in newborn PA but not in 2-week-old animals (Fig. 6A and B).

Fig. 6

Inhibitory effects of the Rho kinase inhibitor Y-27632 on the vasoconstriction induced by U46619 in isolated piglet PA from newborn (NB, A) and 2-week-old piglets (B). The effects of the combinations of Y-27632 plus nifedipine (Nife) or plus calphostin C (Calph) are also shown. Data show mean ± S.E.M. (n=4–11). # and ## P<0.05 and 0.01 Y-27632 vs. control, * P<0.05 vs. Y–27632 alone.

3.6. Effects of vasodilator agents on U46619-induced increase in [Ca²⁺]_i and contraction

In these experiments the effects of NO (5–200 nmol/L), nifedipine (0.1–10 μmol/L), the adenylate cyclase activator forskolin (0.01–1 μmol/L), and Y-27632 (0.1–10 μmol/L) were analyzed after stimulation with U46619 (0.1 μmol/L) in 2-week-old PA loaded with fura 2 (Fig. 7A–D). Nifedipine did not induce a relaxant response although it caused an approximately 80% reduction in [Ca²⁺]_i. On the other hand, NO and forskolin, which essentially act through the increase in cyclic nucleotides (cGMP and cAMP, respectively), completely reversed the increase in [Ca²⁺]_i and induced a partial and complete relaxation, respectively. Finally, Y-27632 almost totally relaxed PA but exerted minor effects on [Ca²⁺]_i. Fig. 7E shows the [Ca²⁺]_i–force relationship for the effects of these vasodilator agents explaining their different profiles. Thus, nifedipine nearly reversed the increase in [Ca²⁺]_i with a minor effect on the contraction, and conversely, Y-27632 abolished the contraction with a minor effect on [Ca²⁺]_i. NO and forskolin showed an intermediate profile, the latter being a more efficient vasodilator.

Fig. 7

Effects of different vasodilator agents on the vasoconstriction and the increase in [Ca²⁺]_i induced by the TXA₂ analog U46619 in 2-week-old piglet PA. (A–D) Concentration-dependent effects of NO, nifedipine, forskolin, and Y-27632 on [Ca²⁺]_i (upper panels) and contractile force (lower panels), respectively. (E) [Ca²⁺]_i-force relationship for the effects of these vasodilator agents. Data show mean ± S.E.M. (n=4–6).

4. Discussion

In the present study, we found that two pathways are involved in TP-receptor induced vasoconstriction in piglet PA. First, the activation of PKCζ leads to K_V channel blockade, membrane depolarization, L-type Ca²⁺ channel activation, and an increase in [Ca²⁺]_i. Second, a RhoA/Rho kinase-dependent pathway which, essentially, does not require changes in [Ca²⁺]_i. The major finding was the different contribution of these two mechanisms during postnatal maturation. The concentration–response curves for the vasoconstrictor effect of the TXA₂ mimetic U46619 were similar in the two age groups. However, when compared to the newborns, in the 2-week-old animals, we found a lower K_V current density, a strong rightward shift of the concentration–response curve to the increase in [Ca²⁺]_i induced by U44619 and a lack of inhibitory effects of the Ca²⁺ channel blocker nifedipine, the broad PKC inhibitor calphostin C or the specific PKCζ inhibitor PKCζ-PI on U46619-induced vasoconstriction. On the other hand, the RhoA/Rho kinase pathway plays a more important role in the older animals, acting as a possible counterbalance mechanism and resulting in no apparent changes with age in either the potency or the efficacy of the vasoconstrictor effects of U46619.

In PA, the resting membrane potential appears to be predominantly regulated by K_V channels [16–18,20,35]. Activation of pulmonary K_V channels leads to hyperpolarization, whereas their inhibition causes membrane depolarization, activation of voltage-gated L-type Ca²⁺ channels, increase in [Ca²⁺]_i and vasoconstriction. Changes in the expression or function of K_V channels in PASMC have been involved in the pathogenesis of primary and anorexigen-induced pulmonary hypertension [17,19–21] and in postnatal vascular development [36]. In the present study, we found that the activation kinetics, voltage–activation curves, and sensitivity to 4-AP were similar in the two age groups suggesting that the same or similar channel proteins underlie K_V currents. However, an increase in membrane capacitance and a marked decrease in K_V current density occurred during postnatal maturation. These data was very consistent with those previously described by Evans et al. [37], which observed a twofold increase in membrane capacitance and a threefold decrease in K_V current density in 2-week-old versus newborns. Interestingly, we found a stronger contractile effect of 4-AP in newborn arteries. This different role of K_V currents in regulating PA contractility is likely to be due to the observed changes in K_V current density. However, we cannot rule out that changes in the expression of L-type Ca²⁺ channels could also contribute to this effect.

In rat PASMC, U46619 induced the translocation of PKCζ from the cytosolic to the membrane fraction [11]. In addition, it inhibited K_V currents, increased [Ca²⁺]_i through L-type Ca²⁺ channels and induced a contractile response and all these effects were inhibited by calphostin C and PKCζ-PI. U46619 had no direct effect on L-type Ca²⁺ currents in voltage-clamped cells indicating that increased Ca²⁺ entry through nifedipine-sensitive pathway is secondary to membrane depolarization [11]. Herein, we demonstrated that in piglet PA PKCζ is also expressed and U46619 inhibits K_V currents. Furthermore, a nifedipine- and PKCζ-PI-sensitive increase in [Ca²⁺]_i was also observed, indicating that the PKCζ-K_V-Ca²⁺ pathway is also operative in the piglet. These results do not fully exclude that other mechanisms, apart from K_V currents blockade, may contribute to the increase in [Ca²⁺]_i induced by U46619. The blockade induced by U46619 seems to be less marked in porcine than that observed in rat PA, which is consistent with a reduced sensitivity to 4-AP in piglet versus rat PASMC K_V channels. The reason for this is unclear but could be related to the presence of different types of K_V channel subunits.

The functional role of the PKCζ-K_V-Ca²⁺ pathway was clearly different in the two age groups, i.e., inhibition of PKCζ or L-type Ca²⁺ channel blockade reduced the vasoconstriction induced by the U46619 in newborn but had no effect in 2-week-old PA. Changes with age in the expression and subcellular distribution of PKCζ, in the total phosphorylated PKCζ and in the percentage of K_V blockade induced by U46619 were minor and not statistically significant. Therefore, in agreement with the different contractile response induced by 4-AP, it seems that the K_V channel blockade induced by U46619 is less effective to induce vasoconstriction in the older animals. Furthermore, the concentration–response curve of the [Ca²⁺]_i increase induced by U46619 was shifted to the right in the older animals, so that the contractile responses to the lower concentrations of U46619 were observed in the absence of changes in [Ca²⁺]_i. Another evidence supporting that changes in [Ca²⁺]_i are not required for U46619-induced contraction is the lack of effect of nifedipine or PKCζ-PI on these contractile responses despite a full inhibition of the increase in [Ca²⁺]_i.

These results pointed out that other mechanisms different from the PKCζ-K_V-Ca²⁺ pathway are responsible for the vasoconstriction induced by TP receptor activation in PA, particularly in the older animals. These mechanisms should be independent of changes in [Ca²⁺]_i, i.e., it implies Ca²⁺-sensitization mechanisms [12,33]. The most plausible candidate was the RhoA/Rho kinase pathway since it induced a Ca²⁺-independent vasoconstriction via phosphorylation and inhibition of myosin light chain phosphatase [22] and has been involved in TXA₂-induced vasoconstriction in pulmonary and systemic arteries [13,29]. In the present study, the Rho kinase inhibitor Y-27632 inhibited U46619-induced contraction in piglet PA at both ages. In newborn PA, the inhibitory effect induced by Y-27632 plus nifedipine was similar to that induced by nifedipine alone. The lack of additive effect of these drugs may suggest that Rho-mediated activity is linked to the nifedipine sensitive mechanism. In fact, it has been reported a Ca²⁺-dependent activation of RhoA/Rho kinase in vascular smooth muscle stimulated with TXA₂[38]. However, the inhibitory effect induced by Y-27632 was much greater and was associated with minor effects on Ca²⁺ levels in 2-week-old PA, which is consistent with the view that Rho kinase essentially operates through changes in the Ca²⁺-sensitization. It has been reported that the expression of RhoA increased in 2-week-old as compared to newborn piglet PA [25], which indicated that the RhoA/Rho kinase pathway is upregulated during the postnatal period and that this upregulation may counteract the reduction in the PKCζ-Kv-Ca²⁺ pathway. Taken together, the present results show that the relative contribution of PKCζ-K_V-Ca²⁺ pathway (which depends on the K_V current density) and the RhoA/Rho kinase pathway (which depends on the level of RhoA expression) could vary during vascular development. Thus, in newborns, which showed larger K_V currents, the PKCζ-K_V-Ca²⁺ pathway seems to be play a major role, whereas in older animals, showing smaller K_V currents and increased expression of RhoA, the RhoA/Rho kinase pathway appears to be the main contributing mechanism. For comparative purposes, U46619-induced contractions in rat PA, which show large K_V currents (current densities about 10-fold of those in 2-week-old piglets), were strongly inhibited by nifedipine and PKC inhibitors but unaffected by Y-27632 [11].

Increased activity of TXA₂ and isoprostanes activating TP receptors is associated with several forms of pulmonary hypertension [2–9]. The present study shows that the signaling pathway for TP receptor activation-induced pulmonary vasoconstriction changes during postnatal maturation. Thus, it is likely that the responses to other pulmonary vasoconstrictors involved in pulmonary hypertension, such as endothelin-1, 5-HT, and angiotensin II, might also be affected by the changes in K_V currents or in the RhoA/Rho kinase pathway. Furthermore, the present results raise important issues regarding the choice of the adequate vasodilator to treat this condition since vasodilator drugs can differently affect [Ca²⁺]_i and Ca²⁺ sensitization. For instance, when the RhoA/Rho kinase pathway is the primary mechanism (as in 2-week-old piglet PA), Ca²⁺ channel blockers are ineffective, while RhoA inhibitors are very effective drugs. NO or NO donors which operate through both [Ca²⁺]_i dependent and independent pathways [34] can fully revert the increase in [Ca²⁺]_i induced by TP-receptor activation but only partially reduce the vasoconstriction [32]. Finally, adenylate cyclase activation are very effective in reducing both the increase in [Ca²⁺]_i and the contractile response. Several mechanisms can account for cyclic nucleotide Ca²⁺-desensitization [22]. In the present conditions, it seems that the cAMP- is more effective than cGMP-pathway to reduce Ca²⁺ sensitivity.

In conclusion, two major signaling pathways are involved in TXA₂-induced pulmonary vasoconstriction in piglet PA, the PKCζ-K_V-Ca²⁺ and the RhoA/Rho kinase signaling pathways, which are downregulated and upregulated, respectively, during postnatal maturation. The reduction of K_V current density and the increase of RhoA expression seem to be responsible for these maturational changes. The different contribution of these pathways, depending on the age, animal species, and the vasoconstrictor agent, could have an important influence on the responsiveness to vasodilator drugs used in the treatment of pulmonary hypertension.

Acknowledgements

This work was supported by Comisión Interministerial de Ciencia y Tecnología (SAF 2002/02304) and Danone/UCM (PR32/04-12719) Grants. A. Cogolludo and L. Moreno are supported by Red Temática de Investigación Cardiovascular and Ministerio de Educación, Cultura y Deporte Grants, respectively.